Price Movements Research Articles

MFB: A Generalized Multimodal Fusion Approach for Bitcoin Price Prediction Using Time-Lagged Sentiment and Indicator Features

Bitcoin’s volatile nature has made its price prediction a sought-after mathematical model in the FinTech industry. Existing studies, however, need to look into the critical aspect of time-lagged sentiment in Bitcoin price forecasting. This omission is significant because time-lagged sentiment captures delayed market reactions that are not immediately apparent in price movements. Moreover, the correlation between time-lagged sentiment and technical indicators and the limitations of individual machine learning and deep learning models necessitates a comprehensive approach for accurate and reliable Bitcoin price predictions. This paper introduces the multimodal fusion Bitcoin (MFB), an innovative generalized multimodal fusion approach that effectively integrates BiLSTM and BiGRU layers for complex feature extraction. The model employs the BorutaShap algorithm for feature selection and utilizes attention mechanisms and spatial dropout for optimization and generalization. MFB’s training and validation use news and tweet data combined with Bitcoin technical indicators to explore the impact of time-lagged sentiment on price movements, leading to more accurate and timely market predictions. The MFB performs superior Bitcoin prediction performance, achieving 97.63% accuracy and an MAE of 0.0065. Experiments highlight MFB’s capability to outperform existing models, offering significant insights for investors in making informed decisions. MFB’s innovative methodology, particularly in next-hour Bitcoin price forecasting, marks an advancement in financial forecasting. By capturing the nuanced dynamics of market sentiment and its delayed effects, MFB is a pioneering multimodal fusion approach in the FinTech domain, revolutionizing Bitcoin price prediction.

Spatial Price Transmission and Dynamic Volatility Spillovers in the Global Grain Markets: A TVP-VAR-Connectedness Approach.

The global food market's escalating volatility has led to a complex network of uncertainty and risk transmission across different grain markets. This study utilizes the Time-Varying Parameter Vector Autoregression (TVP-VAR)-Connectedness approach to analyze the price transmission and volatility dynamics of key grains, including wheat, maize, rice, barley, peanut, soybean, and soybean meal, and their dynamic spillover directions, intensity, and network. By integrating the TVP-VAR-Connectedness model, this research captures the time-varying variability and interconnected nature of global grain price movements. The main findings reveal significant spillover effects, particularly in corn prices, with prices of soybean dominating other grains while prices of peanut and corn experience higher external spillover effects from other grains. The conclusions drawn underscore the imperative for policymakers to consider a holistic perspective of all types of grains when addressing global food security, with this study providing valuable insights for risk management in the grain sector at both global level and country level.

A Stock Prediction Method Based on Multidimensional and Multilevel Feature Dynamic Fusion

Stock price prediction has long been a topic of interest in academia and the financial industry. Numerous factors influence stock prices, such as a company’s performance, industry development, national policies, and other macroeconomic factors. These factors are challenging to quantify, making predicting stock price movements difficult. This paper presents a novel deep neural network framework that leverages the dynamic fusion of multi-dimensional and multi-level features for stock price prediction, which means we utilize fundamental trading data and technical indicators as multi-dimensional data and local and global multi-level information. Firstly, the model dynamically assigns weights to multi-dimensional features of stocks to capture the impact of each feature on stock prices. Next, it applies the Fourier transform to the global features to capture the long-term trends of the global environment and dynamically fuses these with local and global features of the stocks to capture the overall market environment’s impact on individual stocks. Finally, temporal features are captured using an attention layer and an RNN-based model, which incorporates historical price data to forecast future prices. Experiments on stocks from various industries within the Chinese CSI 300 index reveal that the proposed model outperforms traditional methods and other deep learning approaches in terms of stock price prediction. This paper proposes a method that facilitates the dynamic integration of multi-dimensional and multi-level features in an efficient manner and experimental results show that it improves the accuracy of stock price predictions.

Digital Grading Company: Deep Neural Networks for Predictive Analytics for Alternative Asset Portfolio Management

In this study, we investigate the application of deep neural networks for predictive analytics in the realm of alternative asset portfolio management, focusing on Trading Card Games (TCGs) like Pokemon. Our primary objective is to understand the optimal trading cards for investment and to determine the appropriate timing for divestment and retention of these assets. Drawing on the successful implementation of machine learning predictive analytics in traditional financial markets, we aim to extend these techniques to alternative investments. To achieve this, we employ a two-pronged approach using a comprehensive dataset of historical sales values for Pokemon TCGs spanning five years, comprising over 200 million unlabeled data points. The first approach utilizes a semi-supervised sequence learning method, where a Long Short-Term Memory (LSTM) model is initially trained as a sequence autoencoder. The pre-trained model's weights are then used in a supervised phase to fine-tune the predictions for future trading card values. The second approach involves heuristic pretraining, which leverages market-derived indicators from existing literature to bridge the gap between traditional feature engineering and deep learning methods. Our technology stack is designed to handle end-to-end data collection, analytics generation, and execution of recommended trading strategies. Simulated backtesting results demonstrate that our Digital Grading Company (DGC) portfolio model outperforms traditional indices, achieving a 357% return compared to the 61% return of the S&P 500 Index over the same period. Real-time analysis further indicates that our model consistently identifies profitable trading opportunities in the TCG market using hourly data and proprietary algorithmic trade logic, executing trades 24/7. The DGC Predictive Portfolio, based on ungraded Pokemon TCG trading cards, also shows superior performance when compared to the Nasdaq 100 Index (Invesco QQQ ETF). Our analytics framework not only predicts the direction of price movement but also provides insights into the magnitude of changes, enabling a more dynamic and responsive trading strategy. The findings suggest that the integration of deep learning models with heuristic indicators has the potential to revolutionize portfolio management in alternative investments, paving the way for more accurate and profitable decision-making in niche markets. This research highlights the promising role of machine learning in enhancing alternative asset management strategies, suggesting future applications in other asset classes. We conclude that deep neural networks, when combined with data-driven heuristics, can significantly enhance trading outcomes, making them valuable tools for investors in the evolving landscape of alternative investments.

The dynamic relationship between ASEAN+6 exchange rates and stock markets: application of the ARDL model

PurposeThis study investigates the short- and long-run relationship between stock prices and exchange rates in the Association of Southeast Asian Nations (ASEAN)+6 markets. In the short run, support is found for both the theory of a goods market where exchange rates influence stock returns and portfolio balance theory, where stock returns influence exchange rates.Design/methodology/approachThe co-integration approach of linear and nonlinear augmented autoregressive distributed lag (ARDL) models was applied to daily data from 2 January 2017 to 30 June 2023.FindingsThe findings provide evidence that the goods market theory is supported solely in Indonesia and Singapore, while the portfolio balance theory is supported for Australia, China, India and Malaysia.Practical implicationsPolicymakers and investors should seriously consider the importance of the study findings. The results show that all ASEAN+6 countries experience a short-term interaction between the two markets. This illustrates that exchange rates and stock price movements play vital roles in other markets. Stakeholders, particularly policymakers, should be aware of this critical relationship.Originality/valueThis is the first study to dynamically examine how the ASEAN+6 framework influences currency rates and stock markets.

The Market Efficiency Debate: Developments of Financial Market Theories

A united hypothesis to explain the price movement of financial assets has been a significant research goal over the past few decades. This paper reviews the existing literature on both Efficient Markets and Behavioral Finance, concluding with more contemporary literature that attempts to combine the two theories. It summarizes the EMH and Behavioral Finance concepts, the debate between them, and the evidence supporting both. By recognizing the vast swaths of evidence in support of both hypotehses, one can conclude that the questions surrounding the efficiency or inefficiency of markets remain unanswered. Instead, the evidence supporting both hypotheses suggests the necessity of a compromise to rationalize these results instead of relying on one hypothesis or another. The compromise between both hypotheses, in turn, makes the debate between efficiency and inefficiency almost moot. Researchers must account for human misbehavior and rational expectations on a spectrum when formulating new market hypotheses, remaining open to changes in market theory as new variables continue to be discovered.

Determinants JCI of the Indonesian Stock Exchange (Comparison of IHSG Projection of OLS Method and ARIMA and GARCH Methods for the Period 2015-2023)

This study aims to examine the effect of interest rate, exchange rate, inflation, gross domestic product and LQ-45 index variables on the movement of the Composite Stock Price Index using the ordinary least square method. Comparison in predicting the movement of the stock price index through three models, and then compared with the projection results with OLS, ARIMA and GARCH methods. This research was conducted using monthly data on JCI, interest rates, exchange rates, inflation, gross domestic product and the LQ-45 index for the period May 2015 - December 2023. From the results it can be concluded that interest rates have a significant negative effect on JCI, exchange rates have a significant negative effect on JCI, inflation has a significant negative effect on the value of JCI, gross domestic product has a significant negative effect on JCI and the LQ-45 index has a positive effect on JCI. The results of the comparison of JCI projections with 3 models showed that the OLS model offers more accurate results and has the lowest MAPE value in predicting stock price movements.

COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING MODELS USING ARIMA AND NEURAL NETWORK AUTOREGRESSION METHODS

Gold price fluctuations have a significant impact because gold is a haven asset. When financial markets are volatile, investors tend to turn to safer instruments such as gold, so gold price forecasting becomes important in economic uncertainty. The novelty of this research is the comparative analysis of time series forecasting models using ARIMA and the NNAR methods to predict gold price movements specifically applied to gold price data with non-stationary and non-linear characteristics. The aim is to identify the strengths and limitations of ARIMA and NNAR on such data. ARIMA can only be applied to time series data that are already stationary or have been converted to stationary form through differentiation. However, ARIMA may struggle to capture complex non-linear patterns in non-stationary data. Instead, NNAR can handle non-stationary data more effectively by modeling the complex non-linear relationships between input and output variables. In the NNAR model, the lag values of the time series are used as input variables for the neural network. The dataset used is the closing price of gold with 1449 periods from January 2, 2018, to October 5, 2023. The augmented Dickey-Fuller test dataset obtained a p-value = 0.6746, meaning the data is not stationary. The ARIMA(1, 1, 1) model was selected as the gold price forecasting model and outperformed other candidate ARIMA models based on parameter identification and model diagnosis tests. Model performance is evaluated based on the RMSE and MAE values. In this study, the ARIMA(1, 1, 1) model obtained RMSE = 16.20431 and MAE = 11.13958. The NNAR(1, 10) model produces RMSE = 16.10002 and MAE = 11.09360. Based on the RMSE and MAE values, the NNAR(1, 10) model produces better accuracy than the ARIMA(1, 1, 1) model.

Machine Learning Applications in Stock Price Prediction

With economic fluctuations, investors focus on the stock market where scientific investment strategies become increasingly crucial. This article utilizes linear regression to predict stock price trends, achieving an R value close to 1, indicating high accuracy in the model's predictions. Additionally, logistic regression and random forest classification models were employed to predict stock price movements, revealing a superior performance with the random forest model achieving an AUC value of 0.64. By employing a combination of regression and classification algorithms, this study offers a diverse perspective on stock price prediction, highlighting the importance of scientific forecasting in financial markets. The findings demonstrate that integrating multiple predictive models enhances the accuracy and reliability of stock price predictions, providing investors with precise market information to facilitate rational investment decisions. Furthermore, this research contributes valuable insights for future studies in stock market prediction, promoting continuous innovation and development in the theoretical and methodological aspects of this field. The results underscore the potential of advanced machine learning techniques in financial forecasting, offering a robust framework for understanding market dynamics and improving investment strategies. This study serves as a reference point for both academic research and practical applications, driving forward the evolution of predictive modeling in finance.

Global Stock Markets during Covid-19: Did Rationality Prevail?

This study assesses the validity of the Efficient Markets Hypothesis (EMH) during the Covid-19 period by evaluating whether various stock markets around the world accurately predicted economic performance in their respective countries. The underlying premise is that stock prices should discount future company cash flows, and the projection of these cash flows can be proxied by investors' ability to forecast macroeconomic conditions. We assess stock market performance from January 2020 to February 2022 in 16 countries, equally split between developed and emerging markets. While results varied, the study suggests that stock markets generally predicted economic activity during Covid-19. The modeling test indicated that stock markets significantly predicted economic activity in 11 out of 16 countries, particularly in developed markets. Furthermore, incorporating the stock market as a variable, improved economic forecasts in all 16 countries. Unlike other studies that mainly focused on the characteristics of share price movements, this research judges the rationality of stock markets against an external criterion. Ultimately, it suggests that EMH was mostly vindicated during the Covid-19 period.

What happens when the disruptive asset meets the conventional asset? An analysis of cryptocurrency and REIT

ABSTRACT This study investigates the symmetric and asymmetric connectedness of recently developed disruptive assets – cryptocurrency and important conventional assets – Real Estate Investment Trusts (REITs) across various time scales and frequencies. We measure the strength and magnitude of returns and volatility spillover between REITs and cryptocurrency using both time series and machine learning approaches. Our findings support the dominance of conventional assets in the correlation with disruptive assets, demonstrated by REITs price movements’ destabilization of the correlation network. Interestingly, this destabilization is concentrated in bear markets, suggesting that while investors trade both cryptocurrencies and REITs in bull markets, investors are likely to hedge their cryptocurrency investments with REITs or avoid cryptocurrencies in bear markets. Our empirical findings provide implications for both asset allocation and policy making.

MACROECONOMIC ASPECTS IN INFLUENCING STOCK PRICE MOVEMENT ON THE INDONESIAN SHARIA STOCK INDEX 2018-2022

This study aims to analyze the influence of macroeconomic variables and world oil prices in influencing stock price movements on the Indonesian Sharia Stock Index (ISSI) in 2018-2022. In this case, stock price movements can be influenced by several factors, one of which is macroeconomics. The research methodology used in this study is to use a quantitative research approach and secondary data type, with quarterly samples from each year 2018 to 2022 which amounted to. The research model carried out is log linear regression panel data analysis conducted through the chow test and classical assumptions using Eviews 10 tools. Based on the results of this study, it states that partially shows the variables of Inflation, BI Interest Rate Economic Growth (GDP) and Money Supply. While simultaneously all variables did not have a significant effect on the Indonesian Sharia Stock Index (ISSI).

Bayesian machine learning framework for characterizing structural dependency, dynamics, and volatility of cryptocurrency market using potential field theory

Identifying the structural dependence between the cryptocurrencies and predicting market trend are fundamental for effective portfolio management in cryptocurrency trading. In this paper, we present a unified Bayesian machine learning framework based on potential field theory and Gaussian Process to characterize the structural dependency of various cryptocurrencies, using historic price information. The following are our significant contributions: (i) Proposed a novel model for cryptocurrency price movements as a trajectory of a dynamical system governed by a time-varying non-linear potential field. (ii) Developed a Bayesian machine learning framework for inferring the non-linear potential function from observed cryptocurrency prices. (iii) Proposed that attractors and repellers inferred from the potential field are reliable cryptocurrency market indicators, surpassing existing attributes in the literature. (iv) Analysis of cryptocurrency market during various Bitcoin crash durations shows that attractors captured the market trend, volatility, and correlation. In addition, attractors aids explainability and visualization. (v) The proposed cryptocurrency market indicators (attractors and repellers) was used to improve the prediction performance of state-of-art deep learning price prediction models.

Is Dhaka Stock Exchange An Efficient Capital Market? - Evidence Of Weak-Form Efficient Market Hypothesis

This paper uses Random walk hypothesis to test the efficiency of Dhaka Stock Exchange Ltd (DSE), an emerging capital market of Bangladesh. In this paper, with Runs test, Dickey-Fuller Unit root test have been processed and analyzed to observed the behavior of daily return of Dhaka Stock Exchange indices during the study period. The sample includes the daily price indices of all securities listed on the DSE general, DSI (All Share), DSE-30, and Daily indices listed with DSE. As a proxy of movement of individual stock prices, daily closing prices of 30 companies operating in the Banking sector has been analyzed. The results provide evidence that DSE does not follow the random walk model and so DSE is not efficient even in weak form. To improve this capital market, the timely disclosure and dissemination of information to the shareholders and investors on the performance of the listed companies should be given emphasized.

Unlocking ETF price forecasting: Exploring the interconnections with statistical dependence-based graphs and xAI techniques

In the complex landscape of financial markets, accurately predicting Exchange-Traded Fund (ETF) price movements requires advanced methodologies. This research introduces a practical approach that integrates network analysis with graph embeddings, specifically utilizing Node2Vec, to enhance financial prediction models' performance and interpretability. By representing the intricate relationships within financial markets in a lower-dimensional space, we improve the efficiency of AI-driven predictions. A key component of our method is applying the SHAP Explainable AI (xAI) framework, which helps interpret our tree-based models' decision-making process. Using six different tree-based models, our approach delivers accurate predictions while maintaining transparency in model interpretation. This combination of graph embeddings and explainability tools enables stakeholders to understand better the factors influencing financial market behavior, improving decision-making based on AI models.

Deep Learning Based Prediction Model for Crude Palm Oil Prices Using News Sentiment Analysis with Sliding Window

Crude palm oil (CPO) price prediction plays an important role in agricultural economic development. Various economics and agricultural related factors have been used to predict CPO prices. Nevertheless, understanding news sentiment features will also be important in CPO price prediction. This paper proposes a CPO price prediction model to help the plantation organizations in the palm oil sector to successfully anticipate CPO price fluctuations and manage the resources more effectively. The CPO price behavior is non-linear in nature, and thus prediction is very difficult. In this paper, an improved version of recurrent network, Long Short-Term Memory (LSTM) based CPO price prediction model with news sentiment is used to produce an enhanced prediction model. The findings of this study show that the LSTM based forecasting model with news headline sentiment using a six-month sliding window produced the best result in forecasting the CPO price movement compared to other sliding window sizes.

An Optimal Multi-Modal Approach for Stock Market Price Forecasting with Fused Sentiment Analysis for Real Time Data

Abstract: This study presents an innovative fusion-based methodology that integrates real-time stock market technical indicators with news sentiment analysis from financial news feeds to enhance stock selection decisions. The proposed framework employs a Bidirectional Long Short-Term Memory (Bi-LSTM) model for forecasting stock prices and a Deep Neural Network (DNN) used in conjunction with transformer-based model for sentiment classification, both optimized through the incorporation of real-time datasets. To further refine feature selection, Artificial Bee Colony (ABC) and Firefly algorithms are utilized, significantly improving the accuracy of price trend predictions and sentiment analysis. Technical indicators from stock market data are processed using the Bi-LSTM model to predict future stock prices. Concurrently, sentiment data from the Economic Times is pre-processed through Term Frequency-Inverse Document Frequency (TF-IDF) vectorization and pre-trained transformer models to extract key sentiment scores. The ABC algorithm enhances textual feature selection, while the Firefly algorithm reduces skewness in the error distribution of technical indicators, thereby improving the alignment between forecasted and actual stock prices. The fusion mechanism integrates predicted price movements with sentiment analysis to provide actionable insights for stock selection. The decision-making framework classifies stocks based on the consistency between price movements and sentiment: stocks predicted to rise with positive sentiment are labelled as "Sure Selection"; rising prices with negative sentiment are categorized as "Dicey Selection"; declining prices with positive sentiment result in a "Risky Selection"; and negative sentiment with falling prices trigger an "Avoid Selection". The proposed approach achieved a prediction accuracy of 67.00% for the Bi-LSTM model, with a precision of 0.714, recall of 0.695, and an F1 score of 0.677, demonstrating the model's robustness. The mean absolute error (MAE) for the Bi-LSTM model was 4.30, indicating strong predictive performance. The combined use of ABC and Firefly algorithms optimizes both technical and sentiment features, significantly contributing to the overall performance of the prediction model. By fusing multiple data sources and employing advanced optimization techniques, this methodology offers a unique and effective solution for stock market prediction and decision-making

Testing of Semi Strong form of Market Efficiency of Indian Capital Market with Reference to Annual Dividend Announcement

The investment decision is influenced by many factors, of which one such factor is the return. The shareholders may get the return in the form of dividends which affects the share prices. The behavior of stock prices is unpredictable as price movement for different activities will move in different ways. The stock price influence activities can be divided into economic and corporate activities. The impact of economic activities will be more or less the same on all the stock prices, while the impact of corporate action varies from one stock to the other. Dividend payment is one of the essential corporate actions that will impact the behavior of stock prices. This research highlights the impact of dividend payment one the behavior of stock prices and their abnormal returns.

Stock Market Prediction Using Fuzzy Logic and Technical Indicators: A Case Study of Indian Banks

This study presents a decision support system for stock market investors that combines technical analysis with fuzzy logic. The system utilizes four essential technical indicators—Moving Average Convergence/Divergence (MACD), Relative Strength Index (RSI), Stochastic Oscillator (SO), and On-Balance Volume (OBV)—to predict future price movements. These indicators are mapped as inputs into a fuzzy inference system, where fuzzy rules and membership functions are applied to generate trading recommendations, categorized as Bullish, Neutral, or Bearish. Experiments using MATLAB with data from 10 Indian banks demonstrate that the system effectively predicts stock movements and improves investment decisions when integrated with additional market information and investor insights

Prediction of stock prices with automated reinforced learning algorithms

AbstractPredicting stock price movements remains a major challenge in time series analysis. Despite extensive research on various machine learning techniques, few models have consistently achieved success in automated stock trading. One of the main challenges in stock price forecasting is that the optimal model changes over time due to market dynamics. This paper aims to predict stock prices using automated reinforcement learning algorithms and to analyse their efficiency compared with conventional methods. We automate DQN models and their variants, known for their adaptability, by continuously retraining them using recent data to capture market dynamics. We demonstrate that our dynamic models improve the accuracy of predicting the directions of various DAX stocks from 50.00% to approximately 60.00%, compared with conventional methods. Additionally, we conclude that dynamic models should be updated in response to shifts rather than at fixed intervals.